ClimateChat-Corpus

该数据集旨在为大型语言模型提供精细调优，以提升其在气候变化问答任务上的性能。它增强了语言模型利用内部知识来处理气候变化问题的能力。数据集涵盖了多个学科领域的问题，包括气候科学26个、生态学22个、环境科学18个、健康科学20个以及地理学19个。该数据集的任务是针对气候变化问答进行指令调优。

This dataset is designed for fine-tuning large language models (LLMs) to improve their performance on climate change question answering tasks. It enhances the capacity of language models to leverage their internal knowledge when addressing climate change-related questions. The dataset includes questions spanning multiple academic disciplines: 26 from climate science, 22 from ecology, 18 from environmental science, 20 from health science, and 19 from geography. The core task of this dataset is instruction fine-tuning for climate change question answering.